Model Predictions and Performance Monitoring

Agency: Environment, Great Lakes, and Energy

Overview

Once calibrated, a groundwater-flow or fate-and-transport model may be applied to evaluate changes in a number of different hydrogeologic or chemical conditions at environmental contamination facilities. Some of the typical model applications are to predict the change in hydraulic heads or groundwater flow directions as a result of changes in hydraulic stresses (e.g. increases in pumping rates, etc), evaluate the effectiveness of a remedial or corrective action, or estimate the migration pathway and concentrations of contaminants in groundwater. However, errors in the model, even though small, can result in gross errors in solutions projected forward in time. It is for this reason that, in addition to remedy assessment, performance monitoring is required to compare future field conditions with model predictions to assess model error.

A model may be considered part of the facility compliance requirements if specified as part of a Remedial Action Plan (RAP), Corrective Action Plan (CAP), Interim Action Designed to Meet Criteria, or negotiated settlement. However, a model cannot provide verification of remedy effectiveness (e.g., hydraulic containment of a contaminant plume or estimation of the chemical concentration at the point of human or environmental exposure). At best, a model can only provide an estimate of the relative effectiveness of a remedial or corrective action. Verification of actual performance must be demonstrated by the measurement of appropriate field data. Performance monitoring is required as a means of physically measuring the actual behavior of the hydrogeologic system and demonstrating compliance with environmental statutes. This is consistent with the Risk Based Corrective Action (RBCA) process. ASTM guidelines state that Predictive modeling is not used in the RBCA process as a substitute for site-specific verification data" (ASTM Standard E 1739-95 (2002), Appendix X3.4.3).

The degree of performance monitoring required at a facility depends on the conditions or actions that have been simulated and the associated level of risk to the downgradient receptors. With any performance monitoring plan and network, there should be a sufficient number of sampling locations that are properly distributed to verify model simulation results. Monitoring wells that are installed to investigate the possible extent of a contaminant plume often are not appropriately located to monitor the performance of a remedy. For this reason, it is very likely that additional nested monitoring wells (individual wells screened at different vertical depths) will be required to verify remedy performance effectiveness and model simulation results.

Examples of model simulation outcomes and the required elements of an effective performance monitoring plan are contained in the following sections.

Hydraulic Containment

A model simulating effective hydraulic containment of a contaminant plume by a pump-and-treat system, for a given constant pumping rate, would show the following:

  • Simulated hydraulic gradients toward the extraction wells over an area greater than the delineated extent of contamination, and
  • Simulated declining chemical concentrations in monitoring wells located downgradient of the simulated extent of capture shortly after the establishment of the capture zone.

A proper performance monitoring plan to verify model predictions and remedy effectiveness would consist of the following:

  • Monitoring of pumping rates to makes sure that actual pumping rates are equal to, or exceed, those used in the model.
  • Measurement of hydraulic heads in all monitoring wells to show hydraulic gradients toward the extraction wells over an area larger than the delineated extent of contamination. Additional piezometers or monitoring wells must be installed if a sufficient number of wells are not available to measure heads, especially between the extraction wells and the downgradient extent of capture. It is not permissible to use water levels measured in pumping wells for this purpose.
  • Collection of groundwater samples in monitoring wells located beyond the simulated extent of capture. Chemical concentrations in groundwater at these points should show a declining trend with time. Additional monitoring wells must be installed if there is not a sufficient number of monitoring wells properly located immediately beyond the downgradient extent of capture.

Contaminant Removal

Some remedial or corrective actions may include removal of contaminated groundwater to reduce the overall chemical concentrations within the plume. Model simulations of an effective contaminant removal remedy would show the following:

  • An overall declining trend in chemical concentrations within the delineated extent of groundwater contamination, and
  • Declining chemical concentrations at locations beyond the downgradient extent of the zone of contaminant removal, and
  • No increase in chemical concentrations at locations where previous sampling had indicated no detectable or very low detectable concentrations of site-specific chemicals.

A proper performance monitoring plan to verify model predictions and remedy effectiveness would consist of the following:

  • Monitoring of pumping rates to make sure that actual pumping rates are equal to, or exceed, those used in the model.
  • Collection of groundwater samples from the extraction system. Chemical concentrations in extracted groundwater and the mass of chemicals removed by the extraction system should show a declining trend with time.
  • Collection of groundwater samples in monitoring wells located within the delineated extent of the contaminant plume. Overall, the concentrations of chemicals in groundwater should show a declining trend with time.
  • Collection of groundwater samples in monitoring wells located beyond the simulated extent of capture or at a compliance boundary. In locations where groundwater contamination exists, chemical concentrations in groundwater at these points should show a declining trend with time. At the compliance boundary, chemical concentrations in groundwater should not show concentrations that exceed applicable compliance criteria. Additional monitoring wells must be installed if an insufficient number of monitoring wells are located, and evenly distributed, immediately beyond the downgradient extent of contaminant removal.

Natural Attenuation

Model simulations of an effective natural attenuation remedy should show that the contaminant plume is stable or shrinking through the following:

  • Declining concentrations of the chemicals of concern in all monitoring wells located within the delineated extent of groundwater contamination,
  • Declining concentrations in appropriate electron acceptors where degradation is the primary attenuation mechanism,
  • Increasing concentrations of degradation by-products where degradation is the primary attenuation mechanism, and
  • No increase in chemical concentrations in monitoring wells located beyond the delineated extent of the stabilized contaminant.

Facilities at which natural attenuation has been simulated would require extensive monitoring of appropriate chemical parameters and hydraulic heads. A proper performance monitoring plan to verify model predictions and remedy effectiveness would consist of the following:

  • Collection of groundwater samples from all performance monitoring wells located within the contaminant plume. Samples would be analyzed for all appropriate chemicals of concern, degradation by-products, and appropriate field parameters. Chemical monitoring would be required at a sufficient number of locations to evaluate the migration or mass removal of contaminants. Sample results should compare well with simulation results.
  • Collection of groundwater samples from all performance monitoring wells located beyond the downgradient extent of the contaminant plume. Samples would be analyzed for all appropriate chemicals of concern, degradation by-products, and appropriate field parameters. Chemical monitoring would be required at a sufficient number of locations to evaluate the potential for downgradient migration of contaminants. Sample results should compare well with simulations results and show no downgradient migration of chemicals of concern above appropriate compliance criteria.
  • Hydraulic-head measurements would be required to verify groundwater and contaminant migration directions.
  • Additional monitoring wells must be installed if an insufficient number of monitoring wells are located, and evenly distributed, within, and along the migration path of the contaminant plume, and immediately beyond the downgradient extent of the stabilized contaminant plume.
  • Further details on monitoring natural attention are contained in RRD Operational Memorandum No 4, Attachment 8, Monitored Natural Attenuation.

Potential Impact to Downgradient Receptors

Models may be used to show the potential for impact to downgradient receptors such as potable water supply wells or surface water bodies (e.g. lakes and streams). The simulation results would show the following:

  • Concentration distribution between the contaminant source area and the downgradient receptor.
  • The expected concentration at the downgradient receptor.

A properly-designed performance monitoring plan to verify the model would consist of the following:

  • Collection of groundwater samples from all wells located between the contaminant source area and the downgradient receptor. These wells should be located along and perpendicular to the primary migration path of the contaminant plume. Samples would be analyzed for all appropriate chemicals of concerns, pertinent degradation by-products, and appropriate field parameters. Chemical monitoring would be required at a sufficient number of horizontal and vertical locations to evaluate the migration rate, dispersion, or mass removal rate of contaminants. Sample results should compare well with simulation results.
  • Collection of groundwater samples along a compliance boundary upgradient of the receptor. Samples would be analyzed for all appropriate chemicals of concern, pertinent degradation by-products, and appropriate field parameters. Chemical monitoring would be required at a sufficient number of horizontal and vertical locations to evaluate the potential impact to the receptor. Sample results should compare well with simulation results.
  • Hydraulic-head measurements would be required to verify groundwater and contaminant migration directions.
  • Additional monitoring wells must be installed if an insufficient number of monitoring wells are located, and evenly distributed within, and along the migration path of the contaminant plume, and along a compliance boundary upgradient of the receptor.

Impact on Surrounding Hydrology

A model may also be used to simulate the impact of pumping from a groundwater-extraction well on the hydrology of nearby surface water bodies, wetlands, or groundwater levels within the same or adjacent aquifers. Model simulations might show the following:

  • Simulated declines in groundwater levels in the region surrounding the extraction well, and
  • Simulated decrease in groundwater discharge rates to surface-water bodies or wetlands, or an increase in groundwater recharge rates from surface-water bodies or wetlands.

A proper performance monitoring plan to verify model predictions for assessing the impact of a groundwater extraction well would consist of the following:

  • Monitoring of pumping rates to makes sure that actual pumping rates are equal to those used in the model.
  • Measurement of hydraulic heads in all monitoring wells and stage elevations in surface-water bodies or wetlands, if applicable, prior to the beginning of groundwater extraction to show "base-line" water level conditions. Additional piezometers, monitoring wells or staff gages must be installed if there are not a sufficient number of wells or staff gages available to measure water levels on a regional basis, especially between the extraction well(s) and the nearest areas of potential conflict. The areas of potential conflict might be existing groundwater-supply wells, surface-water bodies, or wetlands.
  • Measurement of hydraulic heads in all monitoring wells and stage elevations in surface-water bodies or wetlands, if applicable, after the beginning of groundwater extraction to show the impact of groundwater pumping on regional water levels. These data should be collected on a regular basis, for a sufficiently long time period, to show the long-term impact of development on water levels.
  • It is also beneficial to locate monitoring points in areas beyond the expected zone of influence (ZOI) of the extraction well. Water levels should be measured at these points before and after groundwater pumping has begun. The purpose of these monitoring points is to determine background fluctuations in water levels so that such fluctuations might be "removed" from the performance monitoring data.